A robust principal component analysis-based approach for detection of a stator inter-turn fault in induction motors

Abstract Health condition monitoring of induction motors is important because their vital role and wide us in a variety industries. A stator inter-turn fault (SITF) considered to be the most common electrical failure according statistical studies. In this paper, an algorithm for detection SITF presented. It based on one blind source separation techniques called principal component analysis (PCA). The proposed uses PCA discriminate between faulty components motor current signatures voltage from other components. standard deviation decomposed vectors used as criterion. criterion robust non-fault conditions including quality problems large mechanical load changes well harmonic contaminants supply. addition, with straightforward low computational burden process, method computationally efficient. To evaluate performance method, numbers practical simulation scenarios are considered, results confirm good performance, high degree accuracy, convergence speed method.